Exchange transfusion with cell-free hemoglobin (Hb) permits hematocrit and blood viscosity to be reduced without a proportional decrease in 02 carrying capacity, thereby promoting 02 delivery in hypoperfused states such as stroke and shock. Polymers of Hb appear to be better tolerated in clinical trials than the first generation of cross-linked tetrameric Hb, but there are little data available on the cerebrovascular effects of these commercial polymers. Previously, an Hb polymer without residual cross-linking reagents was developed that does not extravasate or produce hypertension. It normally produces pial arteriolar constriction when viscosity is decreased, and dilation when viscosity is increased, to keep cerebral blood flow (CBF) constant. This pial constriction is mediated by O2-dependent P450 omega-hydroxylase which produces the vasoconstrictor 20-HETE. In addition, two recombinant Hb polymers with different 02 affinities were engineered and unexpectedly indicated that Hb with high 02 affinity was more effective in decreasing infarct volume after stroke. In the present application, the optimal 02 affinity of plasma-based Hb for reducing stroke damage will be determined. Pial arteriolar diameter responses in ischemic border regions will be measured during transfusion to determine 1) if high 02 affinity Hb, which should have less precapillary 02 loss, produces less constriction and better perfusion; 2) if inhibiting omega-hydroxylase activity or administering an endothelin-1 antagonist blocks intraischemic constriction to Hb transfusion, promotes CBF and reduces infarct size; 3) if Hb transfusion produces a delayed upregulation of heme oxygenase (HO) activity that counteracts the initial constriction, and if prior upregulation of HO activity prevents constriction of collateral pial arteries. Both gene deletion and pharmacological procedures will be used. Because of the interest in using Hb solutions in hemorrhagic shock, the effect of resuscitation with a large Hb polymer on the peripheral and cerebral circulation will be investigated. Inhibition of omega-hydroxylase will be tested as a strategy to promote cerebral vasodilation after resuscitation from shock. Therefore, this proposal will provide new mechanistic insights into how cell-free Hb polymers can be used to promote 02 delivery during stroke and shock.